Linking Optic Nerve Sheath Diameters and Glioblastoma Multiforme: A Retrospective Analysis

Size: px

Start display at page:

Download "Linking Optic Nerve Sheath Diameters and Glioblastoma Multiforme: A Retrospective Analysis"

Lesley Ross
5 years ago
Views:

1 Linking Optic Nerve Sheath Diameters and Glioblastoma Multiforme: A Retrospective Analysis Andrew Teh, DO Alysha Vartevan, DO Aswin Kumar, DO Department of Radiology, Larkin Community Hospital/Lake Erie College of Osteopathic Medicine

2 Study Purpose Examine the relationship between optic nerve sheath diameter, increased intracranial pressure, and primary intracranial tumors.

3 Background The optic nerve sheath diameter (ONSD) measurement includes the dural sheath, subarachnoid space and optic nerve. Increased ONSD has long been associated with increased intracranial pressure. Increased intracranial pressure is often associated with trauma. Some of the clinical symptoms in increased ICP include headache, vomiting, and drowsiness. These findings are often nonspecific, and there have been attempts to find alternative methods of diagnosing increased ICP including measuring ONSD. To our knowledge, there have been no studies examining the specificity of increased ONSD. More specifically, there have not been studies examining the relationship between ONSD and primary intracranial tumors to our knowledge. We are examining the relationship between ONSD with T2 MRI imaging and Glioblastoma Multiforme (GBM).

4 Methods After Institutional Review Board (IRB) approved the study, images from 76 patients with GBM were retrospectively examined by two independently trained observers. Images were reviewed in two sets, with the initial set examining 24 patients with an additional 52 patients subsequently added. Three patients were excluded from each data set (total 6 excluded) because of infiltrating masses with indistinct dimensions on their MRI scans. A 1.5 T Siemens magnetom Avanto was used to obtain all MR images. Axial T2 MR imaging was utilized in all patients to measure the ONSD. The right and left ONSD was measured at the insertion to the globe. The largest dimension of the tumor as well as volume was considered a standard measurement utilizing post contrast T1 images. In patients with multiple lesions, only the dimensions of the single largest enhancing lesion was recorded.

Results Total patients examined 76 6 excluded for ill defined enhancement N=70 (total data set), N=21 (initial set), N=49 (second set) Descriptive Statistics Mean Standard

5 Results Total patients examined 76 6 excluded for ill defined enhancement N=70 (total data set), N=21 (initial set), N=49 (second set) Descriptive Statistics Mean Standard Deviation N largestdimension (mm) Volume (cc) ONSD_R_AVG (mm) ONSD_L_AVG (mm)

10 Initial 21 patients (data set #1) When controlling for each of the ONSD sides, the ONSD size was shown to significantly (r=.0489, p=0.024) predict 24% of the variance in the size of the tumor in our initial data set of 21 patients.

11 Expanded data set (total 70 patients)

12 N=70 However, when the full set of 70 patients was examined, the ONSD was not shown to predict with statistical significant the largest tumor diameter.

13 N=70 Our data showed that ONSD was also unable to predict tumor volume with statistical significance in our data set of 70 patients.

15 Discussion One previous study found an increased ONSD on MR imaging was associated with increased intracranial pressure in patients with traumatic brain injuries. The measurement of the ONSD was / mm in patients with increased intracranial pressure and / mm in healthy volunteers. Although increased ONSD has long been associated with increased intracranial pressure, there have been no studies to our knowledge examining the specificity of increased ONSD. Our retrospective analysis found a measurement of 6.4 ± 1.0 mm for the ONSD on the right and 6.5 ± 0.9 mm for the ONSD on the left. Initially, our data suggested (N=21) that the ONSD had a direct relationship with GBM tumor size which was statistically significant. Unfortunately, when we expanded the sample size (N=70) we could no longer achieve statistical significance in correlating ONSD with GBM tumor size or volume.

16 Limitations Small sample size. We initially saw statistical significance between ONSD and GBM tumor size with 21 patients, and subsequently lost statistical significance at 70 patients. A larger data set would be necessary improve statistical power of the study. Tumor size and volume. All patients examined in the study had surgically proven GBM, however not all patients had a single unifocal lesion. Only the largest tumor diameter or volume of the single largest lesion was recorded. Calculating the complete intracranial tumor burden (taking into account all enhancing lesions in multifocal cases) may prove to have a stronger correlation with ONSD.

17 Conclusion Although the sample size was small and a larger patient population is needed to improve statistical power, these results unfortunately do not suggest that ONSD size is strongly associated with tumor size. At least, when considering only the single largest enhancing lesion. As previously discussed, taking into account the complete tumor burden in multifocal cases may show a stronger correlation with ONSD.

TCD IN THE NICU, PICU AND OTHER APPLICATIONS. Dorothy Bulas M.D. Professor of Pediatrics & Radiology Children s National Washington D.C.

TCD IN THE NICU, PICU AND OTHER APPLICATIONS Dorothy Bulas M.D. Professor of Pediatrics & Radiology Children s National Washington D.C. Objectives Recognize normal and abnormal cranial blood flow patterns